The tumor suppressor p53 is subject to both nuclear import and export, and both are fast, energy-dependent and lectin-inhibited.

Human p53 was expressed in E. coli, purified, labeled with fluorescein iodoacetamide (IAF) and characterized for sequence-specific DNA binding and epitope disposition. Injected into the cytoplasm or nuclei of 3T3 cells IAF-p53 was imported into or exported from nuclei within minutes. Import was inhibited by coinjection of the lectin wheat germ agglutinine (WGA). In contrast, the peptide-protein conjugate NLS-HSA carrying the nuclear localization sequence (NLS) of the SV40 T antigen was only imported but not exported. 3T3 polykaryons were injected with IAF-p53 and photo-bleached by Scanning Microphotolysis in such a manner that only a single nucleus per polykaryon remained non-bleached. IAF-p53 was found to migrate rapidly (halftime 10 min) from non-bleached into bleached nuclei, while NLS-HSA did not. In digitonin permeabilized cells IAF-p53 was imported into nuclei. When removed from the medium after nuclear accumulation IAF-p53 was exported from the nuclei. Nuclear import and export of IAF-p53 both were rapid (halftimes of a few minutes, 22 C) and strongly inhibited by WGA or incubation on ice. NLS-HSA was only imported but not exported. We conclude that the nucleocytoplasmic transport of p53, in contrast to that of NLS-HSA, is bidirectional and that transport in both directions is carrier mediated and energy dependent. These results suggest that p53 contains nuclear export signals (NES) in addition to import signals (NLS) and thus open new views on the potential regulation of p53 cellular fractions.

Properties of fluorescently labeled Xenopus lamin A in vivo.

Wild type Xenopus lamin A, a lamin A mutant lacking the carboxy-terminal cysteine (C662-->S), and human vimentin were expressed in bacteria and fluorescently labeled with 5-iodoacetamidofluorescein (5-IAF). In vitro reconstitution experiments and microinjection of both lamins into living cells revealed that they were indistinguishable from the non-fluorescently labeled proteins. When the 5-IAF lamin A was microinjected into the cytoplasm of 3T3 cells it was rapidly transported into the nucleus, giving rise within 1 h to a strong lamina fluorescence, whereas the lamin A mutant formed dotlike intranuclear aggregates. 5-IAF lamin A associated with the nuclear envelope of microinjected 3T3 cells and 5-IAF vimentin which was incorporated into the preexisting vimentin filaments of this cell line, were analyzed by photobleaching employing two different methods, (i) scanning microphotolysis using a modified laser scanning microscope, and (ii) the conventional photobleaching technique in which the integral fluorescence of a single spot was measured by photon counting. A low but significant fluorescence recovery was measured within 10 min for both 5-IAF-labeled intermediate filament proteins, lamin A and vimentin, in bleached areas of the nuclear envelope and the cytoplasmic intermediate filaments, respectively.

Confocal laser scanning analysis of the association of fibulin-2 with fibrillin-1 and fibronectin define different stages of skin regeneration.

The fibulins represent a novel family of extracellular matrix proteins. We report the temporo-spatial expression of fibulin-2 in skin regenerating from keratinocyte autografts. In normal dermis, fibulin-2 was associated with the fibrillin-containing microfibrillar apparatus, except for the portion immediately adjacent to the dermo-epidermal junction. In contrast, early regenerating dermis showed numerous fusiform fibrillin-microfibrils along the basement membrane, whereas fibulin-2 was present in a distinct and separate layer below. Both proteins formed independent fibrillar systems also in the reticular dermis without significant colocalization; however, over time both fibril systems became congruent: after 4 mo there was extensive colocalization of fibulin-2/fibrillin in the reticular dermis, after 17 and 24 mo this also occurred in the papillary dermis. Simultaneous visualization of fibulin-2 and fibronectin revealed an inverse pattern: complete colocalization at 7 d and discordant distribution 17-24 mo after grafting. In particular, the fibrillar fibronectin pattern at early time points changed into a faint granular distribution throughout the dermis and along the subbasement membrane region as in normal skin. Dermal fibroblast cultures showed that fibrillin and fibronectin participated in distinct fibrillar systems; however, fibulin-2 colocalized with either protein. We propose that, in regenerating skin, fibulin-2 is a late component of the cutaneous microfibrillar apparatus with an earlier existence in a fibrillar matrix mediated by fibronectin. This suggests interaction of fibulin-2 with both fibronectin fibrils and fibrillin microfibrils, and is consistent with in vitro binding data.

An optical method for recording the activity of single transporters in membrane patches.

We have previously introduced an optical technique for recording the transport of fluorescent substrates by single membrane transporters. Referred to as optical single-transporter recording (OSTR), the method was restricted to cases in which membrane transporters occurred at extremely small densities, namely at one or a few transporters per cell. Here we describe the extension of OSCR from whole cells harbouring a small number of transporters to small membrane patches containing transporters at normal densities. A technique was developed for firmly attaching cells to isoporous filters, i.e. very thin transparent sheets containing homogeneous populations of cylindrical pores. The flux of fluorescent transport substrates across the tiny membrane pieces spanning the filter pores was measured by scanning microphotolysis, a combination of fluorescence microphotolysis and confocal laser scanning microscopy. The technique was tested by attaching erythrocytes to filters containing pores of 1.2, 2.0 or 3.0 microns diameter. After treating filter-attached erythrocyte membranes with streptolysin O, the transport of the fluorescent protein B-phycoerythrin through single streptolysin O pores was observed. From the flux data the functional radius of the streptolysin O pore was derived to be 12.5 +/- 0.9 nm, in very good agreement with previous electron microscopic estimates. The new technique features a number of unique properties: (i) the size of the membrane patch can be chosen within wide limits according to transporter density, (ii) transport can be recorded on many membrane patches in parallel, (iii) both influx and efflux may be analysed employing either photobleaching of fluorescent or photorelease of caged nonfluorescent substrates, (iv) two or more transport substrates may be monitored simultaneously. The new technique can be used, for instance, for analysing the activity of protein/particle pumps, a membrane transport domain not previously accessible to a single-transporter analysis.

Two-photon scanning microphotolysis for three-dimensional data storage and biological transport measurements.

Scanning microphotolysis is a method that permits the user to select, within the scanning field of a confocal microscope, areas of arbitrary geometry for photobleaching or photoactivation. Two-photon absorption, by contrast, confers on laser scanning microscopy a true spatial selectivity by restricting excitation to very small focal volumes. In the present study the two methods were combined by complementing a laser scanning microscope with both a fast programmable optical switch and a titan sapphire laser. The efficiency and accuracy of fluorescence photobleaching induced by two-photon absorption were determined using fluorescein-containing polyacrylamide gels. At optimal conditions a single scan was sufficient to reduce the gel fluorescence by approximately 40%. Under these conditions the spatial accuracy of photobleaching was 0.5 +/- 0.1 micron in the lateral (x.y) and 3.5 +/- 0.5 micron in the axial (z) direction, without deconvolution accounting for the optical resolution. Deconvolution improved the accuracy values by approximately 30%. The method was applied to write complex three-dimensional patterns into thick gels by successively scanning many closely spaced layers, each according to an individual image mask. Membrane transport was studied in a model tissue consisting of human erythrocyte ghosts carrying large transmembrane pores and packed into three-dimensional arrays. Upon equilibration with a fluorescent transport substrate single ghosts could be selectively photobleached and the influx of fresh transport substrate be monitored. The results suggest that two-photon scanning microphotolysis provides new possibilities for the optical analysis and manipulation of both technical and biological microsystems.

Optical single-channel analysis of the aerolysin pore in erythrocyte membranes.

Scanning microphotolysis (Scamp), a recently developed photobleaching technique, was used to analyze the transport of two small organic anions and one inorganic cation through single pores formed in human erythrocyte membranes by the channel-forming toxin aerolysin secreted by Aeromonas species. The transport rate constants of erythrocyte ghosts carrying a single aerolysin pore were determined to be (1.83 +/- 0.43) x 10(-3) s-1 for Lucifer yellow, (0.33 +/- 0.10) x 10(-3) s-1 for carboxyfluorescein, and (8.20 +/- 2.30) x 10(-3) s-1 for Ca2+. The radius of the aerolysin pore was derived from the rate constants to be 19-23 A, taking steric hindrance and viscous drag into account. The size of the Ca2+ rate constant implies that at physiological extracellular Ca2+ concentrations (> 1 mM) the intracellular Ca2+ concentration would be elevated to the critical level of > 1 microM in much less than a second after formation of a single aerolysin pore in the plasma membrane. Thus changes in the levels of Ca2+ or other critical intracellular components may be more likely to cause cell death than osmotic imbalance.

Carboxy-terminal conversion of profibrillin to fibrillin at a basic site by PACE/furin-like activity required for incorporation in the matrix.

Fibrillin-1, the main component of 10-12 nm microfibrils of the extracellular matrix, is synthesized as profibrillin and proteolytically processed to fibrillin. The putative cleavage site has been mapped to the carboxy-terminal domain of profibrillin-1, between amino acids arginine 2731 and serine 2732, by a spontaneous mutation in this recognition site that prevents profibrillin conversion. This site contains a basic amino acid recognition sequence (R-G-R-K-R-R) for proprotein convertases of the furin/PACE family. In this study, we use a mini-profibrillin protein to confirm the cleavage in the carboxy-terminal domain by both fibroblasts and recombinantly expressed furin/PACE, PACE4, PC1/3 and PC2. Site-directed mutagenesis of amino acids in the consensus recognition motif prevented conversion, thereby identifying the scissile bond and characterizing the basic amino acids required for cleavage. Using a PACE/furin inhibitor, we show that wild-type profibrillin is not incorporated into the extracellular matrix until it is converted to fibrillin. Therefore, profibrillin-1 is the first extracellular matrix protein to be shown to be a substrate for subtilisin-like proteases, and the conversion of profibrillin to fibrillin controls microfibrillogenesis through exclusion of uncleaved profibrillin.